New prediction markets report notes 4x YoY volume growth to $63.5B in 2025 as structural risks remain unresolved and $3.7B in new capital pours into fresh industry startups.
According to the 2026 Skynet Prediction Markets Report by CertiK, a leading Web3 security firm, prediction markets saw annual trading volume grow fourfold in 2025 — from $15.8 billion to $63.5 billion. A small group of dominant platforms now controls the majority of global activity.
Kalshi was valued at $11 billion after closing a $1 billion Series E round led by Paradigm in late 2025, with backing from Sequoia Capital, Andreessen Horowitz, and others. Polymarket reached a $9 billion valuation following a $2 billion strategic investment from Intercontinental Exchange. In total, venture investors deployed $3.7 billion into prediction market startups last year, minting multiple billionaire founders, according to Forbes.
Chemistry co-founder and managing partner Mark Goldberg told Forbes that prediction markets could see trading volumes grow “100x in the next five years,” reflecting investor expectations of exponential expansion.
Consolidation at the top, pressure below
Skynet’s central conclusion is that dominance is concentrating quickly. Kalshi, Polymarket, and Opinion now control the majority of global volume. But each of them is pursuing a distinct model: a federally regulated exchange, a crypto onchain venue, and a hybrid blockchain platform.
But the report is equally focused on fragility. It highlights third-party infrastructure as a systemic weakness, pointing to the December 2025 Magic.link authentication incident that affected Polymarket users. Even when smart contracts remain secure, hybrid Web2/Web3 architectures can introduce centralized failure points.
Prediction markets reached $63.5B in 2025, marking 4x growth. But with scale came new risks: oracle exploits, admin key abuse, and Web2.5 infrastructure issues.
Prediction markets reached $63.5B in 2025, marking 4x growth. But with scale came new risks: oracle exploits, admin key abuse, and Web2.5 infrastructure issues.
How did Kalshi, Polymarket, and Opinion lead the pack?
Read the report to find out 👇https://t.co/YC1KGDKAJe
— CertiK (@CertiK) February 10, 2026
Oracle manipulation remains the primary technical attack vector, according to Skynet, since market resolution mechanisms directly control fund distribution. Administrative key controls and contract upgrade mechanisms persist across many decentralized platforms. This creates potential single points of failure.
At the same time, Skynet notes that wash trading has inflated volume metrics across the industry, with research estimating that artificial activity reached as high as 60% during peak airdrop farming periods. While probability outputs remained largely reliable, distorted liquidity signals could undermine institutional confidence.
Venture capital sees a 100x opportunity
The recent Forbes report detailed how venture capital firms are aggressively backing prediction market startups. Many of them were founded by recent college graduates, with $3.7 billion in fresh capital flowing into the sector in 2025 alone.
“Prediction markets are moving so quickly into the mainstream in a way where I think the amount of money being traded is going to go 100x in the next 5 years,” said Mark Goldberg, cofounder and managing partner of Chemistry, in the Forbes report.
Caitlin Pintavorn, a partner at crypto-focused venture firm Paradigm, told Forbes that prediction markets have been “one of the most popular areas” attracting new founders over the past year. The scale of capital involved reflects the sky-high valuations of Kalshi and Polymarket at the top. Both platforms count Donald Trump Jr. as an investor, adding political visibility to an already high-profile sector.
In January alone, traders wagered more than $10 billion across Kalshi and Polymarket, including $550 million on the Super Bowl, according to the Forbes piece.
VCs Are Throwing Money At Recent College Grads To Build Prediction Markets https://t.co/8xIauYqYEb
(📸: Taraneh Tajdini) pic.twitter.com/iV2l9juInk— Forbes (@Forbes) February 10, 2026
New forms of prediction markets startups emerge
Unlike earlier prediction platforms focused strictly on binary yes/no contracts, many new startups highlighted by Forbes are experimenting with new derivatives and data models.
New York-based Noise, backed by Paradigm, offers so-called “attention markets” where users can go long or short on internet trends rather than discrete event outcomes. “We felt that the binary option question had exhausted itself in a way,” Noise cofounder Gabriel Perez Carafa told Forbes.
Other startups are positioning prediction markets as hedging tools rather than speculative venues. Pluto, founded by 22-year-old Ronit Jain, aims to create markets around GPU costs for AI computing. Jain described his vision as building a financial layer to help AI companies hedge what he calls “arguably the most important commodity of the next 10 years.”
Even AI model testing is entering the mix. Arcada Labs uses prediction markets to evaluate how large language models trade on platforms like Kalshi, turning market performance into a proxy for AI reasoning ability.
Momentum meets structural reality
The venture narrative is one of acceleration and inevitability. As Forbes noted, regulation in the United States, following Kalshi’s legal victory, has given investors confidence that prediction markets can operate as legitimate financial products.
But Skynet’s report introduces a counterweight. Regulatory fragmentation persists globally, with multiple European Union countries banning Polymarket as unauthorized gambling. Even within the US, state-level restrictions threaten to create a patchwork compliance despite federal jurisdiction.
The platforms that survive, Skynet argues, will be those that can maintain liquidity across fragmented jurisdictions, demonstrate institutional-grade security infrastructure, and build revenue models not dependent on incentive-driven volume.
On one side, venture capitalists are betting on 100x growth and throwing millions at 22-year-old founders building the next iteration of prediction markets. On the other hand, infrastructure analysts warn that oracle risk, admin key control, regulatory fragmentation, and artificial volume remain unresolved structural issues.











